DBT Report

[michaelab]

Originally posted by merlin
And again, why no prosecutions if it's all a con? I simply don't understand that

Because, as has been shown, it's pretty difficult to prove it's a con to most people's satisfaction and anyway, I don't think that even in the US they're going to get to the level of suing cable companies because "I spent $2K on this cable and it's no better than a $10 one".

The people who buy expensive cables don't think they're being conned otherwise they wouldn't buy them (or keep them in their systems) - allthough the reluctance to admit you've been conned is a very powerful force that many con artists use to their benefit.

But in the end it's a bit like saying why don't people prosecute artists? "I paid £100K on this work of art and then I find out it's nothing more than a regular pile of bricks".

Michael.

 

[michaelab]

Originally posted by lowrider
I wonder why all of a sudden the cables to be compared have to be equal in R/C/I

That's always been the requirement, it's not new. The thing is, take a £10K Nordost loudspeaker cable, measure the resistance, capacitance and inductance and any electrical engineer will be able to make you a cable for £10 that will have the same electrical properties.

That's really the point of those blind test challenges: show me any cable you like of any cost and I'll make one for less than $50 which has the same electrical properties and we'll see if you can tell them apart. You'd think that the cable companies would be falling over themselves to take up this challenge - the marketing value would be huge. Strangely, none of them have

Michael.

 

[merlin]

Michael,

that's really scraping the barrel if you don't mind me saying so.

to begin with, you yourself have spent considerable sums on Audiophile cables that you now claim make no difference what so ever. You tell us that this is clearly the case, that ABX, DBT etc is neccessary to prove otherwise.

But you do not report the manufacturer to the Trading Standards Authority and claim compensation for your loss (and your continuing stress as the owner of a public forum!)

Would you be so leniant in any other area of life, say if you bought some loft insulation that claimed to improve your heating, but failed miserably, when measured!, to do so.

The analogy with Art is interesting, but I would certainly question whether you would buy a piece if it appeared to you to be a regular pile of bricks. But then, I have yet to see a piece of art advertised claiming to improve anything in your life. They are simply sold as pieces of work by another human being which you can choose to possess or not. Somewhat different from being coerced into buying some magic cables by misleading advertising surely?

 

[merlin]

Originally posted by michaelab
That's always been the requirement, it's not new. The thing is, take a £10K Nordost loudspeaker cable, measure the resistance, capacitance and inductance and any electrical engineer will be able to make you a cable for £10 that will have the same electrical properties.

That's really the point of those blind test challenges: show me any cable you like of any cost and I'll make one for less than $50 which has the same electrical properties and we'll see if you can tell them apart. You'd think that the cable companies would be falling over themselves to take up this challenge - the marketing value would be huge. Strangely, none of them have

Michael.

Oh fantastic :mana:

So now it's yes they do make a difference, but they are over priced and I can make a copy for a lot less, never mind the R&D time and effort that went into finding a cable that provided the ideal set of measurements.

So what is it Michael. Do all cables sound the same. If not why? If these factors change the percieved performance of cables in systems, is it not unreasonable to suggest that they may just have a beneficial effect?

Whould you like another spade mate

 

[sideshowbob]

The analogy with Art is interesting, but I would certainly question whether you would buy a piece if it appeared to you to be a regular pile of bricks. But then, I have yet to see a piece of art advertised claiming to improve anything in your life.

The pile of bricks making up Carl Andre's Equivalent VIII, was, indeed, no more than a regular pile of fire bricks.

The whole point of the sublime, and therefore art, presumably, is to visualise the transcendent, a pretty improving experience I would say. Andre's bricks don't do that for me mind you.

I can recommend the Donald Judd exhibition at Tate Modern, btw. Lots of "ordinary" aluminium and other industrial materials, but seeing it was a minor life-improver:

http://www.tate.org.uk/modern/exhibitions/judd/

-- Ian

 

[sideshowbob]

Oh, and on the cable thing, show me an "audiophile" cable that measures any differently electrically over normal hi-fi lengths from some bit of wire that can be picked up for a couple of quid from Dixons or Maplins and you may be onto something.

-- Ian

 

[michaelab]

merlin, you have an uncanny and annoying skill of willfully mis-understanding my posts and then twisting a few quotes to hang each new attack on

I was merely saying that those cable challenges require two cables that measure the same electrically. I didn't say that I thought the different electrical properties would make them sound different, they don't. The "same electrically" requirement is to stop some smart-arse turning up with a cable that's basically a capacitor (and sounds crap) and walking away with the prize.

As I already said a few posts back, I think you'd need HUGE and difficult to achieve differences in inductance or capacitance to make audible differences - and cables with those extreme numbers would probably sound crap as they would act like a low pass filter.

In addition, as Ian has already said:
"show me an "audiophile" cable that measures any differently electrically over normal hi-fi lengths from some bit of wire that can be picked up for a couple of quid from Dixons or Maplins".

never mind the R&D time and effort that went into finding a cable that provided the ideal set of measurements

Oh please, give me a break. Cable R&D is throwing darts and seeing if where they land looks nice. There is no ideal set of measurements anyway as all audio cables measure basically the same anyway.

I haven't spent "a considerable amount" on cables. The total cost of the cables in my system is probably less than £350. As soon as I have time to buy some "stock" power cords they'll replace the Eupens and that number will go down to £200.

As far as trading standards go don't be ridiculous. All cables meet trading standards requirements because they do what they are supposed to do, which is carry an electrical signal. If people want to spend 10 times as much as they need to on them then that's up to them. I would only have a trading standards case if any cable I bought didn't conduct electricity.

Michael.

 

[lowrider]

I have a dozen standard PCs, you want them for your Eupens, as they sound the same, I am not paying more... :latte:

 

[michaelab]

Originally posted by lowrider
I have a dozen standard PCs, you want them for your Eupens, as they sound the same, I am not paying more

I actually think I might make some DIY power cables along the lines of the TNT Merlino in honour of my forum jousting partner

It has one advantage over stock power cords in that it uses shielding (which does make a difference) which I don't think most stock leads have. Will mean I can get some quality IEC and mains plugs because if anything is likely to make a difference it's in the connections and not the cable itself.

Michael.

 

[merlin]

Originally posted by michaelab
As far as trading standards go don't be ridiculous. All cables meet trading standards requirements because they do what they are supposed to do, which is carry an electrical signal. If people want to spend 10 times as much as they need to on them then that's up to them. I would only have a trading standards case if any cable I bought didn't conduct electricity.

Michael.

Regardless of the advertising message given out by the manufacturer

 

[sideshowbob]

Merlin, old fruit, you've been known to attack hi-fi manufacturers who make gear that, in your opinion, doesn't perform as advertised. Why don't you take them all to court?

-- Ian

 

[bottleneck]

Originally posted by 7_V


I'm thinking of the sonic differences between two different capacitors which, presumably, have the same capacitance.

Fascinating parallel that Steve, Ive been thinking about it off and on for the past hour.

To paraphrase the question:

If a change of a capacitor (for an alternative higher specification equivalent of equal capacitance) can change the sound, why cant changing a cable of equal capacitance yet different construction equally change sound?

I'd confess that my knowledge of physics, and electronics/electronic components isnt good enough to have a crack at that.

My best hypothesis is that dearer higher quality Caps effect a change in capacitance more uniformly and accurately than a less expensive/lower quality alternative.

Perhaps this quality/performance issue is more applicable to resistors and capacitors than to cable?

Comments welcome

 

[merlin]

Originally posted by sideshowbob
Merlin, old fruit, you've been known to attack hi-fi manufacturers who make gear that, in your opinion, doesn't perform as advertised. Why don't you take them all to court?

-- Ian

Specifics Ian ?

 

[wadia-miester]

Here he comes into the frey battle axe weilding, charger riding and general man about dbt. Heres..................................................................................................................Woolfie

 

[wolfgang]

Merlin,
The Stereophile article is interesting. Never the less the author comes across as having an agenda and it is to prove that ABX/DBT listening tests are not useful. I think I am beginning to understand why some think like this too. Below is taken from page 3 of the article.

''Dr. Corey then inadvertently indicted the proponents of double-blind ABX testing by suggesting that "the experimenters tend to get the results they expect to obtain." He also revealed his bias in suggesting that "six guesses out of seven is not significant," an apparent reference to the many skilled listeners who have made six out of seven correct identifications in double-blind tests. Note the use of the word "guesses" rather than "identifications."

Some people seem to have difficulty with this little point. Unfortunately that statement is absolutely correct.

Look at it this way. When you ask your friend to throw a coin 7 times and he get 6 heads in a row would you think he has special powers? You will say it was due to chance. True?

He then throws 70 times and gets 60 heads. You will still think it could be chance. It is still a reasonable assumption but some people may disagree.

He then throws 700 times and gets 600 heads. Now everyone will ask him to try it again with another coin since the most likely explanation is the coin is funny.

He uses 20 other coins and each time he get at less 80-90% heads. Would you now think he has a special power? I think some will accept he is good at this game but other would still think no.

What am I getting at?

If a few Zerogain members get together to try a Single or Double Listening test with speaker cables, whatever the result I will certainly never expect it will be the last of the Great Debate. Everyone will go away and draw a different conclusion. It is only to be expected.

All we could hope for is it will be fun. IF it is interesting enough then we will discuss it in a sensible manner and plan for further 'improved' listening tests. Maybe just maybe one day we will all come to a unifying UNDERSTANDING whatever it may be.

 

[merlin]

Originally posted by wolfgang
Merlin,
Look at it this way. When you ask your friend to throw a coin 7 times and he get 6 heads in a row would you think he has special powers? You will say it was due to chance. True?

Wofgang,

I find that a poor analogy I'm afraid. When tossing the coin, the result is purely down to chance unless using a rigged coin. If you are talking about guessing heads or tails, again you are talking about pure guesswork. When the coin is in the air, you are not asking the participent to declare heads or tails by using any of the five senses, purely by guessing or using some sixth sense.

By using this analogy, you are inferring that the listeners in question, when asked to identify something using one of their senses, rely instead on pure guesswork. This would statistically not result in an 83% success rate. What you are implying is that the listeners are not using their ears.

Now the requirement to continue the test effectively ad infinatum takes no account of the fact that the skill required to identify the cables is that of listening to music, not that of concentrating on spotting differences for the sake of it. It is also entirely feasible that lengthy comparisons lead to a dulling of the senses as tiredness comes into play.

What you are saying is that if a difference is detected initially, keep testing long enough and it is likely that the results will even themselves out. This takes no account of any changes in the concentration and tiredness levels surely. What would your interpretation be of results that went say 7/7, 6/7, 5/7 , 4/7 , 3/7 etc ending up averaging out as a null? Would that prove that there was no difference, or that fatigue had a large part to play in the result?

 

[wadia-miester]

Originally posted by merlin

Now the requirement to continue the test effectively ad infinatum takes no account of the fact that the skill required to identify the cables is that of listening to music, not that of concentrating on spotting differences for the sake of it. It is also entirely feasible that lengthy comparisons lead to a dulling of the senses as tiredness comes into play.

Merlin, certainly has identified one area that generates inconsistanices, this was identified early on in our testing procedures, after repeating the test (same cables) in succession over first a 9 hour period, then a 5 day period, the cables that were correctly identified in the former tests conducted, were not repeated in the latter part of the day/week.
The testing now used, is limited to 3 single sessions.

 

[lowrider]

It is my experience in our kit tests, if we switch them more than 6 or so times, we begin confusing them, even if the difference was significant on the first tests, that goes for sound or image testing...

 

[wolfgang]

What I am trying to say, on the surface, if some one seems to have manage to correctly identify a cable with DBT 6 out of 7 tries most of us may accept he could be showing an ability to audibly distinguish the 2 cables.

However, a tosser may point out that person could have simply guess all of it and still come out with that result. Mathematically it is true. We cannot be sure.

However, if that person do a bigger number tries and, for example 60 out of 70 right, everyone would be less likely to think it is purely guess work. After awhile if he was guessing the chance should become more even. Or he could be getting tired and unable to tell anymore. So far so good?

The other part is this. Some seem to think if we can agree on one ultimate DBT it would solve the debate. No it would not since it is more likely to uncover some problem we never thought of before the test. Only further tests would make the result more conclusive.

I am just giving fore warning. If a SBT or DBT was ever get organised whatever the result not everyone will have the same conclusion. That is life. We learn a step at a time.

 

[michaelab]

merlin, wolfgang's analogy wasn't so much an analogy as trying to explain some statistics, and giving an example of why statistics is often counter intuitive.

The point is, for a DBT result to be statistically significant you have to do enough trials so that you can say that there's a very low probability (the usual figure used is 5%) that the same results could have happened purely by chance.

If I toss a coin 7 times and get 6 heads the result doesn't mean anything as it's quite likely to happen by chance alone. There's a relatively high probability (much higher than 5%) that in any given 7 coin throws you will get 6 heads (or 6 tails).

By using this analogy, you are inferring that the listeners in question, when asked to identify something using one of their senses, rely instead on pure guesswork. This would statistically not result in an 83% success rate

Oh yes, it could well do. That's the whole point. With a number of trials as low as 7 you would often get an 83% success rate (6 out of 7 btw is closer to 86% but that's not important) even if the listenters were just tossing a coin to make their choice.

The key to a statistically significant test is to do enough trials so that you can say that there is less than a 5% chance that these results could have happened by chance. That requires many trials.

Listener fatigue is a problem though and is one reason why ABX testing with enough trials to be statistically significant is tricky. The obvious solution is to increase the number of listeners. Having 100 listeners over 7 trials is statistically equivalent to having 1 listener do 700 trials. One issue I have with listener fatigue though is that it can really only be a problem when we're talking about very subtle differences. I'm sure that anyone could compare a Sony beat-box to even a modest separates system and choose correctly every time, even after 100 attempts.

The irony is that the more subtle the difference, the more trials are likely to be required to show a statisically significant difference but it's precisely those subtle differences that are most subject to listener fatigue. ABX tests are not perfect that's for sure.

Michael.